factors influencing prolonged stay in the intensive care

NATIONAL JOURNAL OF MEDICAL RESEARCH print ISSN: 2249 4995│eISSN: 2277 8810

NJMR│Volume 5│Issue 2│Apr – Jun 2015 40

ORIGINAL ARTICLE

FACTORS INFLUENCING PROLONGED STAY IN THEINTENSIVE CARE UNIT AFTER CARDIAC SURGERYRitesh Shah1, Ramesh Patel1, Ramanand Sharma2, Himani Pandya3, Komal Shah4

Author’s Affiliations: 1Associate Professor, Dept of Cardiac Anesthesia; 2Research Fellow; 3Research Associate;4Research Officer, Department of Research, U. N. Mehta Institute of Cardiology and Research Center, AhmedabadCorrespondence: Dr. Ritesh Shah, Email: [email protected]

ABSTRACT

Background: There are different risk factors that affect the intensive care unit (ICU) stay after cardiacsurgery. The aim of this study was to evaluate possible risk factors influencing prolonged ICU stay.

Methodology: We conducted 443 adult patients undergoing different cardiac surgery to determinatecauses of prolonged ICU stay. These patients were divided into two groups according to ICU stay ≤ 4days and >4 days. We evaluated preoperative, intraoperative and postoperative risk factors for prolongICU stay.

Results: Among the 443 patients studied, 277 (62.52%) had stayed ≤ 4 days and 166 (37.42%) had stayed>4 days in ICU. Frequency of prolonged ICU stay was 42.7% in patients undergoing coronary artery by-pass graft (CABG), 37.3% in patients with valve surgery, 9% in patients with CABG plus valve surgeryand 10.8% others surgery. Patients with > 4 days of ICU stay received more blood transfusion and higherinotropes duration. They also had longer time duration of surgery, cardiopulmonary bypass, and long ven-tilation hours and other preoperative and postoperative risk factors. (P < 0.05 for all comparisons).

Conclusions: The relationships between the pre-, intra- and postoperative risk factors and prolongedICU stay after coronary artery bypass grafting with cardiopulmonary bypass, suggest that serumcreatinine, urea and liver dysfunction were risk factors for prolong ICU stay, Whereas among surgicalvariables cardio pulmonary bypass time, cross clamp time and time duration of surgery are the main asso-ciates of higher ICU duration.

Keywords: Cardio Pulmonary Bypass, LVEF, Serum Creatinine

INTRODUCTIONAfter a cardiac operation, Intensive care admissionis a standard of treatment for most of the patientsbut, long stay in intensive care unit (ICU) is usuallyassociated with increased hospital mortality, poorlong-term prognosis, and increased morbid-ity.1Many patients with prolonged hospital stayexpired shortly after discharge and that the func-tional status of those who survived for a longerperiod was worse when compared with those whorecovered faster. The prediction of their ICU stayis a fact of great importance, based on the need forhaving the expanding costs involved in cardiacoperations and bed availability, saving resourcesand ICU costs concerning the population of car-diac surgery patients.2 Intensive care requires onlythe use of exclusive equipment, highly expert and

devoted healthcare staff. Intrinsically the ICU takesa major amount of the total healthcare cost andconsequently patients with prolonged ICU stay canhave solemn cost implications. Furthermore, pa-tients with prolonged ICU stay can also lead to ashortage of ICU beds and result in operations be-ing cancelled.

We aimed to investigate the postoperative, preop-erative and intraoperative predictors and prognos-tic significance or prolonged ICU stay amongstcardiac surgery patients. This identification couldlead to the primary recognition of the high-riskpatients for longer ICU hospitalization; also pro-vide an important aid to nursing supervisors andleaders for the suitable reserve planning and actualchoice of the operation list.


NJMR│Volume 5│Issue 2│Apr – Jun 2015 Page 141

METHODOLOGYThis is a prospective observational study con-ducted in U.N. Mehta Institute of Cardiology andResearch Centre from January 2014 to March2014. The study was approved by institute ethicscommitteeand informed and written consent wasobtained from all the patients. Four hundred fortythree consecutive patients who underwent variouscardiac surgical procedures and admitted in to thecardiac surgery ICU were enrolled. The age groupof the patients defined in the study were ≥ 18years to which includes both sexes and a minimumICU stay of 24 hours. The patients transferring toanother ward or hospital within the first two daysof postoperative stay in ICU and death during op-eration were excluded from the study. The lengthof ICU stay after surgery was consideredto be thenumber of days spent in cardiac ICU unit immedi-ately after surgery and 24 hours in the ICU wasmeasured as one day. For the patient who was atthe end of the day readmitted to the ICU, the totalnumber of days (first admission and readmission)was considered. The prolonged stay in the ICUwas defined as stay of >4 days.

The median of ICU stay was 4 days. On the basesof this we had two groups, Group A patients ICUstay ≤4 days and Group B patients ICU stay >4days. All the relevant pre-operative details such asage, sex, pre-operative diagnosis, types of surgeryand history of patient for example diabetes, smok-ing were recorded. Blood investigations were doneon the day before surgery. Echocardiography wasperformed the day before the surgery and all rele-vant details were recorded.

All surgeries were conducted according to the in-stitute protocols for surgery and anaesthesia. Therelevant details such as surgery performed, dura-tion of surgery, cardio pulmonary bypass (CPB)time, aortic cross clamp time and blood transfu-sion details were recorded. Blood transfusion wasdone according to the institute protocol. The basicmonitoring parameters such as arterial pressure,central venous pressure, oxygen saturation, urineoutput etc. were collected for all patients. Echo-cardiography was performed the day after the sur-gery and all complication such as renal, neurologi-cal and sepsis were recorded. Duration of inotropein hours, mechanical ventilation in days was re-corded. Blood investigations were done on firstpost-operative day and subsequently as required.

Statistical Analysis

Statistical analysis was carried out using SPSS ver-sion 20.0 software (SPSS Inc, USA). The chi-squared test and independent sample t-test wereused to compare categorical and continuous vari-ables respectively. Data were presented as mean ±SD or proportion as appropriate. The “p” valueless than 0.05 was considered to be significant.

RESULTSDemographic and preoperative characteristics ofthe sample are shown in Table 1 and 2 respec-tively. The majority of patients were males (Male68.4% and female 31.6%) and the median age was50 years. The study population had undergone ei-ther coronary artery bypass grafting (46.04%) orvalve surgery repair or replacement (35.66%) orcombination of them. The preoperative factorsinvolved in longer and shorter ICU stay are pre-sented in Table 2.

Table 1: Demographic characteristic of thestudy populationVariables Group-A

(N= 277)Group-B(N=166)

pvalue

Age(year) 48.511 ± 13.05 50.52 ± 13.53 0.122Sex M= 194 (70.0%)

F= 83 (29.96%)M=109 (65.6%)F= 57 (34.3%)

0.041

Weight(kg) 57.97 ± 13.46 57.46 ± 16.47 0.723Height (cm) 161.33 ± 13.01 159.49 ± 13.74 0.159BMI 25.98 ± 61.40 25.91 ± 37.42 0.989DM 43 (15.52%) 35 (21.08%) 0.174HTN 20 (7.2%) 14 (8.4%) 0.779Group-A -ICU Stay ≤4 Days; Group-B -ICU Stay >4 Days;All figure are in mean±SD; NS: Non-significant, BMI: Bodymass index, DM: Diabetes Mellitus, HTN: Hypertension

The preoperative level of serum creatinine, SGOT,urea, RBS, albumin and LVEF were found to besignificantly (p<0.05) higher in Group B as com-pare to Group A patients. Intraoperative character-istic of study population are shown in Table 2,same as postoperative data are shown in Table 3.The time duration of surgery is higher in group Band statistically significant (Group A3.88 ± 1.24Group B 4.60 ± 1.59, p<0.001). The cardio pul-monary bypass time and cross clamp time werehigh in group B and statically significant (GroupA49.38 ± 57.03 Group B 73.19 ± 70.03, p=0.003and Group A 37.142 ± 45.57 Group B 50.33 ±50.97, p=0.006).



Table 2: Preoperative findingVariables Group-A

(mean±SD)Group-B(mean±SD)

p Value

SGPT (mg/dl) 28.11 ± 29.70 31.01 ± 32.49 0.337SGOT(mg/dl) 25.49 ± 17.90 31.93 ± 33.59 0.009Total bilirubin (mg/dl) 0.88 ± 0.60 1.0 ± 0.85 0.083Total protein (g/dl) 6.83 ± 0.62 6.826 ± 0.82 0.953RBS(mg/dl) 137.63 ± 64.19 158.47 ± 78.27 0.003Urea (mg/dl) 26.34 ± 11.95 34.26 ± 24.31 <0.001Creatinine (mg/dl) 0.903 ± 0.46 1.079 ± 1.17 0.027Sodium (mmol/litre) 136.67 ± 3.41 136.21 ± 14.63 0.616Potassium (mmol/litre) 4.00 ± 0.50 4.104 ± 0.490 0.033Albumin (g/dl) 3.81 ± 0.51 3.68 ± 0.44 0.006Globulin (g/dl) 3.032 ± 0.51 3.109 ± 0.581 0.145LVEF (%) 49.42 ± 11.80 46.09 ± 13.14 0.010Group-A -ICU Stay ≤4 Days; Group-B -ICU Stay >4 Days; NS: Non-significant, SGOT: Serum glutamic oxaloacetic transami-nase, SGPT:Serum glutamic pyruvic transaminase, RBS: Random blood sugar, LVEF: Left ventricular ejection fraction

Table 3: Intraoperative findingsVariable Group-A (mean±SD) Group-B (mean±SD) p ValueCPB time (min) 49.38 ± 57.03 73.19 ± 70.03 0.003AOX time (min) 37.142 ± 45.57 50.33 ± 50.97 0.006Time duration of surgery (Hours) 3.88 ± 1.24 4.60 ± 1.59 <0.001Group-A -ICU Stay ≤4 Days; Group-B -ICU Stay >4 Days; CPB: Cardio pulmonary bypass, AOX: Aortic cross clamp

Table 4: Postoperative findingsVariables Group-A (mean±SD) Group-B (mean±SD) p ValueSGOT (mg/dl) 56.42 ± 35.88 72.04 ± 60.19 0.001SGPT (mg/dl) 34.32 ± 27.10 38.89 ± 40.76 0.1575Urea (mg/dl) 33.98 ± 13.80 41.84 ± 24.79 <0.001Creatinine (mg/dl) 1.055 ± 0.49 1.138 ± 0.507 0.0892Total Protein (g/dl) 5.820 ± 0.75 5.44 ± 0.93 0.0043Albumin (g/dl) 3.190 ± 0.418 2.93 ± 0.55 <0.001Total Bilirubin (mg/dl) 1.442 ± 1.045 1.846 ± 1.66 0.0021RBS (mg/dl) 203.56 ± 61.88 250.16 ± 85.70 0.0040Total WBC(/cmm) 17268.26 ± 8935.53 15960.47 ± 5750.56 0.0922Platelets (/cmm) 215221.42 ± 103564.944 187508.85 ± 82733.03 0.0044LVEF (%) 49.423 ± 11.80 46.095 ± 13.14 0.0102Inotropes Duration (Hours) 29.94 ± 20.10 86 ± 72.874 <0.001Ventilations (Hours) 8.198 ± 6.723 31.92 ± 52.88 <0.001Post op Hospital Stay (Days) 7.761 ± 3.049 11.15 ± 6.95 <0.001Mortality (%) 2.8 12.04 0.0003Group-A -ICU Stay ≤4 Days; Group-B -ICU Stay >4 Days; SGOT: Serum glutamic oxaloacetic transaminase, SGPT:Serumglutamic pyruvic transaminase, RBS: Random blood sugar, LVEF: Left ventricular ejection fraction, WBC: White blood cell

Postoperative complications of both the groupsare shown in Table 4. Similar to preoperative find-ings the higher level of SGOT, urea, RBS, albuminand lower LVEF were associated with Group Bpatients. Apart from this, other morbidities such ashigher protein loss, higher level of bilirubin,greater inotrope duration, ventilation and postop-erative hospital stay was observed with group Bpatients which was statistically significant the inci-

dence of mortality were also higher in Group B(20%) patients as compare to Group A (8%).

In Table 5 summarizes the main findings of thecorrelation analysis. A correlation was measuredprior to conducting the multivariate logistic regres-sion analysis to check the independent variables.Because duration of cross clamp time, cardiopul-monary bypass were and preoperative and postop-erative LVEF strongly correlated (P < 0.001) withprolong ICU stay.



Table 5: Independent correlation with prolongICU stayVariables Pearson Correlation p ValueGroup-APre op SGOT (mg/dl) -.119* 0.049Pre op LVEF (%) -.168** 0.009Post op RBS (mg/dl) .195* 0.011Post op LVEF (%) -.168** 0.009CPB time (min) -.159** 0.009AOX time (min) -.155* 0.011Group-BPost op LVEF (%) -.314** <0.001Post op RBS (mg/dl) .376** <0.001CPB time (min) -.255** 0.001AOX time (min) -.290** <0.001Group-A -ICU Stay ≤4 Days; Group-B -ICU Stay >4 Days;*Correlation is significant at the 0.05 level (2-tailed); **. Corre-lation is significant at the 0.01 level (2-tailed).SGOT: Serum glutamic oxaloacetic transaminase,SGPT:Serum glutamic pyruvic transaminase, RBS: Randomblood sugar, LVEF: Left ventricular ejection fraction,CPB:Cardio pulmonary bypass, AOX: Aortic cross clamp

The group wise surgeries are shown in Table 6.The frequencies of on pump and off pump surger-ies are shown in Table 7.

Table 6: Types of surgeryVariable Group-A

(n=277)Group-B(n=166)

CABG (%) 48 42.77Valvular (%) 34 37.34CABG+ Valvular (%) 3.9 9Other (%) 13.35 10.84Group-A -ICU Stay ≤4 Days; Group-B -ICU Stay >4 Days;CABG: Coronary artery bypass grafting

Table 7: The frequencies of on pump and offpump surgeriesVariable Group-A (n=277) Group-B (n=166)On pump (%) 51.98 37.34Off pump (%) 48.01 62.65Group-A -ICU Stay ≤4 Days; Group-B -ICU Stay >4 Days;

DISCUSSIONSeveral studies have been done to determine therisk factors which may lead to the prolong stay inICU after cardiac surgeries. Multifactorial natureand wide range of patients’ conditions make theunderstanding of individual factor’s contributiondifficult. Any of these factors can affect other fac-tors and strengthen or weaken their effect. Themore complicated the surgery is, the higher therate of mortality, morbidity and postoperative

complications .The main findings of our studywere the significant association of four preopera-tive parameter such as serum creatinine, SGOT,and urea with the prolonged ICU stay of cardiacsurgery patients. In particular, patients with preop-erative renal dysfunction, high perioperative risk,and high mean glucose levels intraoperative havegreater possibility to stay in the cardiac surgeryICU more than 4 days.

The preoperatively increased serum creatinine lev-els were significantly higher with prolonged ICUstay. In concordance to our findings, several otherreports have concluded that the history of renaldysfunction is an independent predictor for pro-longed hospitalization into the intensive care unit.3,

4, 5, 6 In addition, Ranucci et al. 7, in a retrospectivestudy of 9120 cardiac surgery patients revealed theincreased preoperative creatinine levels among themain determinants of late discharge from the ICU.

In our study we have also found out postoperativeRBS, urea, SGOT, LVEF and inotrope durationand prolong ventilation hours significantly corre-lated with prolong ICU stay. Among the postop-erative and intraoperative factors, units of trans-fused blood and other blood products in operationroom, administer of inotropes and its duration inafter surgery, duration of CPB, and time durationof cross clamp time affected the duration of ICUstay.

Most studies reported prolonged (> 72 hours) ICUstay rate of about 33% while we faced with thesame rate by defining prolonged ICU stay as stay >96 hours. For example, in study of Hein et al. ICUstay was longer (> 72 hours) in 26% of cardiacsurgery patients.8 In another study by Mahesh et al.prolonged ICU stay (> 72 hours) rate was 18.7%.9Type of operations and their proportions mightplay roll in ICU stay time. Only 42.7% of cases inour study underwent CABG, whereas in most cen-tres (even in our centre) about 70% of the adultcardiac surgeries are CABG, which might affectthe study result and prolonged ICU stay. Durationof ventilation and inotrope administration in ICUcan be considered as independent predictors ofprolonged ICU stay in our study.

CONCLUSIONThe relationships between the pre-, intra- andpostoperative risk factors and prolonged ICU stayafter coronary artery bypass grafting with cardio-pulmonary bypass, suggest that serum creatinine,urea and liver dysfunction were risk factors forprolong ICU stay, Whereas among surgical vari-



ables cardio pulmonary bypass time, cross clamptime and time duration of surgery are the mainassociates of higher ICU duration. These findingspoint to the necessary evaluation of strategies forreduction of hospitalization time and suggests spe-cial attention to certain clinical and/or surgicalconditions that may, when prevented or bettercontrolled, reduce the length of hospital stay forpatients undergoing coronary artery bypass graft-ing.

REFERENCES1. Jan O Friedrich,Gail Wilson and Clarence Chant. Long-

term outcomes and clinical predictors of hospital mortalityin very long stay intensive care unit patients: a cohortstudy. Crit Care. 2006; 10(2):1-9.

2. B. Ray, D. P. Samaddar, S. K. Todi, N. Ramakrishnan,George John, and Suresh Ramasubban. Quality indicatorsfor ICU: ISCCM guidelines for ICUs in India. Indian JCrit Care Med 2009; 13(4): 173–206.

3. C. Herman, W. Karolak, A. M. Yip, K. J. Buth, A. Hassan,and J. F. L´egar`e. Predicting prolonged intensive care unitlength of stay in patients undergoing coronary artery by-pass surgery—development of an entirely preoperativescorecard.Interact CardiovascThoracSurg 2009; 9(4):654-8

4. R. Atoui, F. Ma, Y. Langlois, and J. F. Morin.Risk factorsfor prolonged stay in the intensive care unit and on theward after cardiac surgery. Journal of Cardiac Surgery2008; 23(2): 99–106.

5. Jan Bucerius, Jan F. Gummert, Thomas Walther, NicolasDoll, Volkmar Falk, Dierk V. Schmitt, et al., Predictors ofprolonged ICU stay after on-pump versus off-pump coro-nary artery bypass grafting . Intensive Care Medicine 2004;30(1):88-95.

6. S. V. Ghotkar, A. D. Grayson, B. M. Fabri, W. C. Dihmis,and D. M. Pullan. Preoperative calculation of risk for pro-longed artery bypass surgery. Kidney International 1999;55(3): 1057–1062.

7. M. Ranucci, C. Bellucci, D. Conti, A. Cazzaniga, and B.Maugeri. Determinants of early discharge from the inten-sive care unit after cardiac operations.Ann ThoracSurg2007; 83:1089–95.

8. Hein OV, Birnbaum J, Wernecke K, England M, KonertzW, Spies C. Prolonged intensive care unit stay in cardiacsurgery: risk factors and long-term-survival. Ann Thorac-Surg2006; 81(3):880–5.

9. Mahesh B, Choong CK, Goldsmith K, Gerrard C, NashefSA, Vuylsteke A. Prolonged stay in intensive care unit is apowerful predictor of adverse outcomes after cardiac op-erations. Ann ThoracSurg 2012; 94(1):109–16.

factors influencing prolonged stay in the intensive care

Documents